# D type flip flop without clock

Is it possible to make a D type flip flop without a clock input? How would one set the logic to 0 or 1 without using a clock input? Thank you.

• That's a S-R (or R-S) Latch. Look it up.
– user16324
Mar 7, 2020 at 13:27
• Thank you, I’ll have a look
– QMD
Mar 7, 2020 at 13:29

## 2 Answers

By my definition of "flip-flop" and my definition of "clock", the answer is no.

By my definition, a flip-flop is an edge-triggered bistatic element. The signal that provides the triggering edge is conventionally called the clock.

• Thanks for the quick response. Do you think it would be possible if the output Q or ‘Q was fed into the clock input? Or maybe adding an inverter?
– QMD
Mar 7, 2020 at 13:27
• No, I don't think that would be possible. If you add some kind of feedback you may end up with a bizarre oscillator, which is no longer a "D flip-flop". Mar 7, 2020 at 13:28
• Ok, thanks for your help
– QMD
Mar 7, 2020 at 13:29

A reasonable question... Congratulations for the insight! I always ask my students when considering flip-flops, "Is it possible to make an asynchronous D flip-flop?" I have also asked this question in the RG forum. Here is my explanation why it is not possible...

Bistable circuits with memory (latches, flip-flops, RAM cells...) are toggled by two different input signals - to set the circuit to "1" and "0". They must be applied only for a short time... and then the input sources must be disconnected... like in S&H circuit.

Another life analogy is the ordinary toggle switch. Once you press it (to turn the lamp on or off) then you have to release it, right?

This can be made in two ways - by two inputs or by one input and clock. The former is implemented in asynchronous RS latches where there is a "neutral" input combination (two ones for NAND latches and two zeros for NOR latches). The latter is implemented in synchronous D flip-flops by level or edge-triggered clock.

Another well-known analogy is the electric switch for lighting. Once you press it (to turn the lamp on or off) then you have to release it, right?

So, the role of the clock is to provide momentarilly acting input signals. In the case of the asynchronous D flip-flop, there is no "neutral" input state when the input source is disconnected. So it is not possible to make a D type flip-flop without a clock input.

"Do you think it would be possible if the output Q or ‘Q was fed into the clock input?"

A similar way is to make T flip-flop from D flip-flop - by connecting the inverted output to D input and using the clock as an input. However, the new T flip-flop is asyncronous.

• No, the input sources do not need to be disconnected. Do you mean deasserted? There is no requirement that they be "applied only for a short time". Where I live, "ordinary light switch" is a toggle switch and you can push it in either direction indefinitely; there is no need to release it. Mar 7, 2020 at 17:57
• I mean that after the input signal is applied to the latch input (S) it has to be removed to allow the latch to be driven by the other input (R). If you know (or can remember), in old transistor latches it was done by connecting RC differentiating circuits before inputs. After you pushed the switch (ON), you have to take your finger off so that someone else can switch it (OFF); otherwise, it will be blocked. The toggle switch possesses memory; it is a push button with memory... it is a mechanical latch... so there is no need to hold it pushed... Aren't these simple concepts really clear? Mar 7, 2020 at 18:33
• Your words...disconnect...remove...suggest a physical severing of a connection. I think you mean that the logical state must change to a non-asserted state. Regarding the light switch, I didn't say that you "need to hold it pushed", I said there is no need to release it. You said "once you press it...you have to release it" which, taken as you said it, is not true. The difference in these phrases is important, and if you use the wrong language then no, it is not really clear. Mar 7, 2020 at 18:45
• The language barrier appears from time to time... but I think the idea of ​​my explanation has finally become clear. The sample&hold analogy is excellent for this case. We apply "1" and strobe the input to sample it; the latch holds it. Then we apply "0" and strobe the input again to sample it; the latch holds the new content... Mar 7, 2020 at 19:08